Clinical Cancer Research Molecular Diagnostics in Cancer Therapeutic Development: Fulfilling the Promise of Personalized Medicine Infection and Cancer: Biology, Therapeutics, and Prevention
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
1078-0432.CCR-05-1117v1
12/19/5641    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Curran, E. M.
Right arrow Articles by Estes, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Curran, E. M.
Right arrow Articles by Estes, D. M.
Clinical Cancer Research Vol. 12, 5641-5647, October 1, 2006
© 2006 American Association for Cancer Research

Human Cancer Biology

Estrogenic Regulation of Host Immunity against an Estrogen Receptor–Negative Human Breast Cancer

Edward M. Curran1,2, Barbara M. Judy1,2, Ngozi A. Duru1,2, Hui-Qun Wang3, Leoncio A. Vergara4, Dennis B. Lubahn5,6 and D. Mark Estes1,2

Authors' Affiliations: 1 Department of Pediatrics, 2 Sealy Center for Vaccine Development, 3 Sealy Center for Environmental Health and Medicine, and 4 Department of Neuroscience and Cell Biology, University of Texas Medical Branch at Galveston, Galveston, Texas and Departments of 5 Biochemistry and 6 Child Health, University of Missouri-Columbia, Columbia, Missouri

Requests for reprints: Edward M. Curran, Department of Pediatrics, Sealy Center for Vaccine Development, 2.330G Children's Hospital, Galveston, TX 77555-0372. Phone: 409-772-0435; Fax: 409-772-0460; E-mail: emcurran{at}utmb.edu.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: The risk of developing breast cancer is positively correlated with exposure to increased levels of estrogen and/or an increased duration of estrogen exposure. Many different mechanisms have been proposed to explain the association of estrogens with breast cancer risk; however, the well-documented immune modulatory properties of estrogen have received little attention. In part, this is due to a lack of suitable models for studying this relationship.

Experimental Design: We have developed an animal model using estrogen receptor (ER)-negative human breast cancer cell line, MDA-MB-468, xenografted into severe combined immunodeficient (SCID) mice. We also generated the ER-{alpha} knockout (ER-{alpha}KO) mice on the SCID background and then tested the ability of 17ß-estradiol to stimulate growth of xenografted ER-negative human breast cancer tumors in wild-type and ER-{alpha}KO SCID mice. We quantified vascularization of tumors, macrophage recruitment to the tumor site by immunocytochemistry, and inflammatory cytokine production.

Results: We show that estrogen treatment of C57BL/6/SCID mice promotes the growth of xenografted ER-negative tumors in wild-type mice and this estrogen-induced tumor growth is abrogated in ER-{alpha}KO mice. Tumor neovascularization of estrogen-treated mice was unchanged versus control; however, estrogen treatment of the C57BL/6/SCID host suppressed macrophage recruitment to and inflammatory cytokine production at the tumor site.

Conclusions: These data are consistent with estrogen modulation of the inflammatory response as a contributing factor in estrogen-stimulated growth of an ER-negative tumor. This effect on the host innate immune response was mediated by ER-{alpha}.

There is a significant body of literature that links exposure to estrogen as a risk factor for the development of breast, ovarian, endometrial, and prostate cancers (reviewed in refs. 1, 2). Postulated mechanisms by which a "natural" hormone, such as estrogen, may promote cancer development include the following: direct stimulation of proliferation in breast tissue, increasing the probability of introduction of a harmful mutation in dividing cells (3, 4); indirectly, by induction of other growth factors and growth factor receptors (5); and abnormal exposure during critical periods of development, direct stimulation of estrogen receptor (ER)-positive tumor cells, and metabolism of 17ß-estradiol (E2) to compounds that are capable of inducing DNA damage (reviewed in ref. 6). E2 modulation of immune cells involved in the surveillance and subsequent destruction of breast and other cancers has not been fully elucidated. Steroids, such as estrogen, progesterone, and glucocorticoids, have well-documented effects on immune function. Expression of ERs has been reported in thymocytes and thymic epithelial cells, spleen, CD8+ suppressor/cytotoxic T cells, and murine peripheral blood monocytes and macrophages (reviewed in refs. 7–9). Murine natural killer (NK) cells (10) and human neutrophils (11) express ER-{alpha} and ER-ß. We have also shown by reverse transcription-PCR with primers specific for ER-{alpha} and ER-ß that murine macrophages and dendritic cells express only ER-{alpha} (12). ER-{alpha} expression has also been shown in follicular dendritic cells present in germinal centers at or near neoplastic lesions in human breast cancer patients (13). The widespread expression of ER-{alpha} in cells of the immune system supports the idea that ER is necessary for the development and maintenance of normal immune function.

Modulation of innate immune effector cell activity, including NK, macrophage, and neutrophil lytic pathways, represents potential targets for therapeutic intervention in breast and other cancers. Clearly, activation (or prevention of inhibition) of NK cells, macrophages, dendritic cells, and/or neutrophils against tumor targets with mild or no side effects would be beneficial. For example, many studies have shown that in vivo treatment of mice with E2 inhibits NK-mediated cytotoxicity, suggesting that endocrine therapies using selective ER modulators, such as tamoxifen that inhibit the action of estrogen, may stimulate the innate immune response to breast cancer. Antiestrogens, such as tamoxifen, modulate innate immune functions, including cytotoxicity in murine NK cells (14–17) and in human NK cells, from patients with breast cancer (18, 19). These studies suggest that tamoxifen treatment could benefit ER-negative as well as ER-positive breast cancer patients. However, tamoxifen treatment has proven ineffective for treatment of ER-negative breast cancers (reviewed in ref. 20). Aromatase inhibitors, such as anastrozole or letrozole, have not been used in ER-negative breast cancer patients and may prove more effective in these patients. Model systems that can discern the effect of host immune components on tumor growth are needed. Discerning the effect of in vivo estrogen or antiestrogen treatment on host immune function is difficult to resolve with ER-positive breast cancer xenograft models where both the tumor and host immune system express ERs. Friedl and Jordan (21) described an ER-negative cell line, MDA-MB-231 subclone 10A, which exhibited estrogen-responsive growth when xenografted into E2-treated nude mice. In this model, E2 also stimulated tumor growth in beige nude mice, which lack NK cells, ruling out the possibility that E2-induced suppression of NK cellular cytotoxicity mediated the E2-induced growth of tumors. However, other aspects of the immune response to tumor formation were not tested by these investigators.

We have developed a model in which E2 treatment of tumor-resistant C57BL/6/severe combined immunodeficient (SCID) mice (which lack functional T and B cells) enhances growth of MDA-MB-468 tumors implanted in these animals. Furthermore, estrogen-stimulated growth was abrogated in ER-{alpha} knockout (ER-{alpha}KO) mice. Direct stimulation of proliferation through MDA-MB-468 is largely minimal because MDA-MB-468 cells lack ER and do not respond to in vitro E2 treatment. We support the hypothesis that E2 treatment enhances MDA-MB-468 tumor growth in SCID mice by suppression of innate immune function with data indicating that the immune response is, at least in part, compromised in E2-treated mice. This model provides a rational basis for the evaluation of estrogen action on the innate immune response to breast cancer.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Animals and surgical procedures. SCID mice [C57BL/6J-Prkdcscid/SzJ (C57BL/6/SCID)] were purchased from The Jackson Laboratory (Bar Harbor, ME). C57BL/6 mice carrying the disrupted ER-{alpha} coding sequence were crossed with C57BL/6/SCID mice, and C57BL/6/SCID mice carrying the disrupted ER-{alpha} coding sequence (ER+/ER–) were maintained as a breeding colony at the University of Texas Medical Branch (Galveston, TX). Wild-type (ER+/ER+) and ER-{alpha}KO (ER–/ER–) animals used in these experiments were either littermates or age matched whenever possible. All experiments were conducted according to the University of Missouri-Columbia (Columbia, MO) or the University of Texas Medical Branch Animal Care and Use guidelines as recommended by NIH. For ovariectomy, animals were anesthetized with 4% isofluorane and bilateral incisions were made in the abdominal cavity as described previously (10). Each ovary was removed, the wound closed, and each animal received 0.75 mg buprenorphine HCl (Reckitt Benckiser Pharmaceuticals, Inc., Richmond, VA) for analgesic purposes. The animals were allowed to recover for 10 days and then divided into two or three groups; the first group received a E2 (0.25 mg/21-day release or 0.18 mg/90-day release) pellet (Innovative Research, Sarasota, FL) by incision with a 12-gauge trocar needle; the second group was subjected to the same surgical procedure, but no pellet was implanted (sham control); and the third group was subjected to the same surgical procedure with no pellet implanted followed by injection with 100 µL anti-asialoGM1 (WAKO, Waco, TX) to deplete NK cells. E2 treatment was initiated 3 days before tumor implantation. Anti-asialoGM1 (100 µL) was injected 1 day before, the day of, and 1 day after implantation of MDA-MB-468 tumor cells.

Cell culture and tumor implantation. MDA-MB-468 cells were maintained in MEM supplemented with 10% fetal bovine serum (Hyclone, Logan, UT), 2 mmol/L L-glutamine, 1x nonessential amino acids (Sigma Chemical, St. Louis, MO), 100 units/mL penicillin G, and 100 µg/mL streptomycin (MEMc) at 37°C and humidified air with 5% CO2. To remove cells from T-150 culture vessels, each flask was rinsed in 15 mL Ca2+-free, Mg2+-free Hanks buffer with 1 mmol/L EDTA (CMFH) and incubated in 35 mL CMFH at 37°C until cells were visibly detached. Cells were rinsed twice in MEMc, and 5 x 106 MDA-MB-468 cells were implanted s.c. in the right axillary mammary fat pad of each mouse. Mice were sacrificed at 8 to 12 weeks of age by CO2 inhalation as recommended by NIH guidelines. Tumor volume was determined by measurement with vernier calipers using two similar formulas: for cylindrical tumors, measurements were made in three perpendicular directions with tumor volume = length x width x height; for spherical tumors, measurements were made in two perpendicular directions with tumor volume = width2 x length. If more than one tumor mass existed, the volume of each mass was determined separately and the volumes were added to give the total tumor volume. After determination of tumor volume, tumors were sectioned and aliquots were either fixed in 10% formalin or snap frozen in liquid nitrogen and stored at –80°C. Untreated mice served as negative controls and anti-asialoGM1-treated mice as positive controls for tumor growth in a NK-deficient environment.

Proliferation assay. For proliferation assay, cells were seeded at 2,000 per well in a 96-well plate in phenol red–free MEM containing 5% charcoal-stripped fetal bovine serum as described previously (22). The cells were cultured in estrogen-free medium for 3 days and incubated in medium alone or medium containing 10–9 mol/L E2, 10–7 mol/L ICI 182,780 (Tocris Chemical, St. Louis, MO), or 10–9 mol/L E2 plus 10–7 mol/L ICI 182,780 as indicated. Medium was changed daily until day 7 when the wells (four wells per treatment) were washed with HBSS and assayed for DNA content using the diphenylamine assay adapted for a 96-well tissue culture plate format (23).

Immunocytochemistry. Buffered formalin-fixed, paraffin-embedded tissue sections (5 µm) were deparaffinized and rehydrated by passage through xylene and graded ethanol solutions. Slides were then treated with 3% hydrogen peroxide with 0.03% sodium azide in PBS for 10 minutes followed by microwave antigen retrieval at 100°C for 10 minutes in DAKO Target Retrieval Solution (DAKO Corp., Carpinteria, CA) in a H2800 Microwave Processor (Energy Beam Sciences, Inc., Agawam, MA). Following sequential 15-minute incubations with 0.1% avidin and 0.01% biotin (Vector Laboratories, Inc., Burlingame, CA) to block endogenous avidin and biotin, slides were incubated in 0.05% casein (Sigma Chemical)/0.05% Tween 20 (DAKO)/PBS for 30 minutes to block nonspecific protein binding. Rat anti-mouse F4/80 monoclonal antibody CI:A3-1 (Abcam, Inc., Cambridge, MA) was applied to sections as indicated at a 1:100 dilution for 60 minutes. Rat serum, negative control, Ready-to-Use (InnoGenex, San Ramon, CA) was used as a negative control. Biotinylated affinity-purified goat anti-rat IgG (Vector Laboratories) that served as the secondary antibody was detected by streptavidin-horseradish peroxidase (HRP) and colorized by 3,3'-diaminobenzidine (DAKO). Primary rabbit anti-human factor VIII antibody (Sigma) was applied at 1:2,000. Universal rabbit negative control (DAKO) was used as negative control as recommended by the manufacturer. DAKO EnVision+ System (HRP-labeled polymer anti-rabbit) was used for detection, and slides were subsequently counterstained with Mayer's Modified Hematoxylin (Poly Scientific, Bay Shore, NY).

Image analysis. All sections were imaged using bright-field microscopy on a Nikon Eclipse 800 upright microscope equipped with a Nikon DXM 12000 digital color camera (Nikon Instruments, Melville, NY). F4/80 and Factor VIII staining was quantified using MetaMorph image analysis software version 6.0 (Universal Imaging Corp., Downingtown, PA) on digital images acquired with a 20x Plan Fluor 0.5 numerical aperture objective. On each digital field, the pixels positive for the HRP staining product were detected using the MetaMorph color thresholding function, which differentiates background nonspecific staining from bona fide staining. The number of positive pixels was determined in 20 different fields from four to five different tumors for each treatment. These measurements were normalized to the total number of pixels per field or, in some cases, to the number of pixels within a specified region of interest selected using the MetaMorph region of interest drawing tool. The normalized data correspond to the percentage of tumor area positive for the HRP staining product. This procedure was used to estimate the relative number of intratumoral macrophages (F4/80 staining) and the degree of vascularity (Factor VIII staining) of each tumor.

The relative number of macrophages on the periphery of each tumor was determined using a method similar to the one described above. To maximize the area per field, digital images of F4/80-stained sections were acquired using a 2x Plan Fluor objective. Due to the unevenness of illumination inherent to the low magnification fields, all images were preprocessed using the MetaMorph shading correction function. When a single x2 magnification field was not enough to cover the entire tumor, an extended view was created by "stitching" several overlapping fields using the "photomerge" function of Photoshop Elements version 3.0. The relative number of peripheral macrophages was estimated by measuring total F4/80 staining in the tumor periphery using the MetaMorph image analysis functions in an analogous manner to the procedure described above for the x20 magnification sections. Macrophage infiltration was estimated as the area positive for F4/80 staining and expressed as a percentage of the total live tumor tissue estimated as the area positive for hematoxylin staining.

Cytokine assays. Frozen tumor specimens from control (n = 4) and E2-treated (n = 4) mice were thawed and homogenized in lysing buffer (150 mmol/L NaCl, 15 mmol/L Tris base, 1 mmol/L CaCl2, 1 mmol/L MgCl2, 0.5% Triton X-100). Homogenates were centrifuged at 1,000 x g for 15 minutes, and supernatants were analyzed with the Bio-Plex Mouse 18-Plex Assay (Bio-Rad, Hercules, CA), which detects up to 18 individual cytokine and chemokine analytes per sample as indicated. Cytokine and chemokines tested included interleukin (IL)-1ß, IL-2, IL-4, IL-5, IL-10, granulocyte macrophage colony-stimulating factor (GM-CSF), IFN-{gamma}, tumor necrosis factor-{alpha}, IL-1{alpha}, IL-3, IL-6, IL-12 (p40), IL-12 (p70), IL-17, G-CSF, KC, macrophage inflammatory protein 1{alpha}, and RANTES.

Statistical analysis. When more than two groups were analyzed, mean tumor volumes from each group were evaluated statistically by one-way ANOVA with Dunnett's test for multiple comparisons against control or median tumor volumes were analyzed by a Kruskal-Wallis test with a Dunn's multiple comparison test using GraphPad Prism software (GraphPad Software, Inc., San Diego, CA). When two groups were analyzed, differences in median tumor volumes from each group were evaluated statistically using a Mann-Whitney rank sum test. Factor VIII and F4/80 staining analyzed at x20 magnification is reported as the mean of 20 different fields from four to five different tumors. Peripheral F4/80 staining analyzed at x2 magnification is reported as the mean of the total staining of each tumor in a group. Differences in the mean between two treatment groups were evaluated statistically using a Mann-Whitney rank sum test. A P < 0.05 was considered significant in all cases.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
E2 stimulates ER-negative MDA-MB-468 tumor growth. MDA-MB-468 cells were xenografted in the right axillary mammary fat pad of intact C57BL/6/SCID mice. Tumors were resected and measured 8 to 12 weeks after implantation. Treatment of intact mice with 0.25 mg/21-day release pellets beginning 3 days before implantation of MDA-MB-468 cells resulted in significantly larger tumors when compared with tumors in untreated control mice (Fig. 1A ; P < 0.01). Figure 1A illustrates the results of a single experiment in which estrogen-stimulated growth was observed. C57BL/6/SCID mice are relatively resistant to xenografted tumor cell lines compared with other mouse strains (24); therefore, as a control we depleted NK cells from intact C57BL/6/SCID mice by injection of anti-asialoGM1 antibodies before implantation of MDA-MB-468 cells. The resultant tumors were also significantly larger than tumors in control mice after 8 to 12 weeks of growth (Fig. 1A; P < 0.05). Implantation of 0.25 mg/21-day release E2 pellets yields a constant high serum E2 level (~800 pg/mL) compared with normal, cycling C57BL/6 mouse E2 levels (10-25 pg/mL; ref. 10). The comparison of tumor growth in control versus E2-treated mice was done in eight separate experiments, and estrogen-stimulated growth was observed in seven of eight experiments, although in some experiments 0.18 mg/90-day release E2 pellets were used in place of 0.25 mg/21-day release E2 pellets (0.18 mg/90-day release pellets yield serum E2 levels of ~400 pg/mL). Figure 1B summarizes the results from all eight experiments. Median tumor volumes from E2-treated and anti-asialoGM1-treated mice were both significantly larger than tumors from control mice (P < 0.05).


Figure 1
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 1. E2 stimulates ER-negative MDA-MB-468 tumor growth. A, intact C57BL/6/SCID mice were divided into three groups; the first group was implanted with a E2 (0.25 mg/21-day release) pellet 3 days before tumor implantation; the second group was subjected to the same surgical procedure, but no pellet was implanted; and the third group was subjected to the same surgical procedure with no pellet implanted followed by injection with 100 µL anti-asialoGM1 (aaGM1) 1 day before, the day of, and 1 day after implantation of MDA-MB-468 tumor cells to deplete NK cells. Columns, mean; bars, SE. P was calculated using a one-way ANOVA with a Dunnett's multiple comparison test. *, P < 0.01; **, P < 0.05. B, summation of eight separate experiments. Point, single animal. Horizontal lines, median tumor volume. P was calculated using a Kruskal-Wallis test with a Dunn's multiple comparison test. *, P < 0.05; **, P < 0.05.

 
In contrast to E2-treated mice, intact control mice were likely subject to hormonal fluctuations due to the estrous cycle. Because immune functions, such as NK cytotoxicity (25), have been reported to fluctuate during the estrous cycle of mice, we wanted to rule out the possibility that differences in tumor growth between E2-treated and control mice were due to cycling by control mice. Therefore, we ovariectomized all animals and then treated one group with an 0.18 mg/90-day release E2 pellets. E2 treatment of ovariectomized mice also resulted in significantly larger tumors compared with control mice (data not shown), indicating that estrous cyclical fluctuations were not involved in the effect of E2 treatment on tumor growth.

ER-{alpha} mediates E2-stimulated ER-negative MDA-MB-468 tumor growth. To determine whether E2-stimulated growth of ER-negative MDA-MB-468 tumor xenografts was mediated by ER-{alpha}, we crossed mice carrying the disrupted ER-{alpha} gene with C57BL/6/SCID mice to generate ER-{alpha}KO SCID mice. MDA-MB-468 cells were xenografted in the right axillary mammary fat pad of intact ER-{alpha}KO C57BL/6/SCID mice. Treatment of intact ER-{alpha}KO C57BL/6/SCID mice with 0.18 mg/21-day release E2 pellets beginning 3 days before implantation of MDA-MB-468 cells did not result in significantly larger tumors when compared with tumors in untreated control mice (Fig. 2 ; P > 0.05). These data are consistent with a functional role for ER-{alpha} expressed in innate immune cells, including macrophages, dendritic cells, and NK cells.


Figure 2
View larger version (6K):
[in this window]
[in a new window]
  		Fig. 2. ER-negative MDA-MB-468 tumor growth in ER-{alpha}KO mice. C57BL/6/SCID/ER-{alpha}KO mice were implanted with a E2 (0.18 mg/90-day release) pellet 3 days before tumor implantation. MDA-MB-468 cells (5 x 106) were implanted s.c. in the right axillary mammary fat pad of each mouse. Mice were sacrificed at 8 to 12 weeks of age, and tumor volume was determined as in Fig. 1. Horizontal lines, median tumor volume. No significant differences in median values were observed as determined by the Mann-Whitney test (P = 0.810).

 
E2 has been shown to stimulate antiapoptotic Akt signaling pathway in ER-negative human breast cancer cell lines (26). To ensure that E2 did not stimulate tumor growth by acting directly on the MDA-MB-468 cells through a non-ER-mediated pathway, we show that MDA-MB-468 proliferation in vitro was not increased in the presence of 10–9 mol/L E2 (Fig. 3 ).


Figure 3
View larger version (16K):
[in this window]
[in a new window]
  		Fig. 3. E2 does not stimulate ER-negative MDA-MB-468 tumor growth in vitro. Cells (2,000) per well were seeded in each well of a 96-well plate in phenol red–free MEM containing 5% charcoal-stripped fetal bovine serum. The cells were cultured in estrogen-free medium for 3 days and incubated in medium alone or medium containing either 10–9 mol/L E2, 10–7 mol/L ICI 182,780, or 10–9 mol/L E2 + 10–7 mol/L ICI 182,780 for 4 additional days. DNA content was determined using the diphenylamine assay adapted for a 96-well tissue culture plate format. There were no significant differences as calculated by one-way ANOVA.

 
E2 does not alter tumor angiogenesis. E2 has been reported to exert a positive influence on tumor angiogenesis in mouse models of mammary cancer and to be a negative prognostic factor for human breast cancer (27, 28). Intratumoral vascular endothelial growth factor levels are altered by E2 treatment in vivo (29). Because these E2-induced increases in angiogenesis were correlated with increased tumor size in these models, we postulated that E2 stimulation of angiogenesis may be a factor in greater tumor growth observed in E2-treated animals. To test this hypothesis, we compared total Factor VIII (von Willebrand factor) staining in tumors from E2-treated versus control mice. There were no detectable differences in Factor VIII staining in control (Fig. 4A ) versus E2-treated (Fig. 4B) tumor sections. Cumulative staining for Factor VIII from 20 independent sections from each treatment group is shown in Fig. 4C (P > 0.05). These data do not support the hypothesis that E2 stimulation of angiogenesis was a contributing factor in the increased tumor sizes observed in E2-treated mice.


Figure 4
View larger version (32K):
[in this window]
[in a new window]
  		Fig. 4. Vascularity of tumors grown in control versus E2-treated mice. Bright-field photomicrographs (x20 magnification) of tumor sections from control (A) and E2-treated (B) mice stained with anti-Factor VIII antibody. Bar, 1 µm. Sections were counterstained with hematoxylin. Arrows, large vessels staining positive for Factor VIII. To estimate the vascularity of each tumor section, the pixels positive for the HRP staining product were detected using the MetaMorph color thresholding function and normalized to the total number of pixels per field determined in 20 different regions from four to five different tumors for each treatment. C, mean Factor VIII staining of 20 control and 20 E2-treated sections. Columns, mean; bars, SE. No significant differences in median values were observed as determined by the Mann-Whitney test. P > 0.05.

 
E2 decreases the number of tumor-associated macrophages. Tumor-associated macrophages (TAM) can either inhibit or promote tumor growth. Inhibition of tumor growth can occur via secretion of immunomodulating cytokines, presentation of tumor antigens to CD4+ T cells, and by direct macrophage-mediated cytotoxicity against tumor cells (reviewed in ref. 30). TAMs coopted by the tumor can function to potentiate tumor growth and progression to a more aggressive phenotype (30–32). TAMs have been shown to produce cytokines, angiogenic factors, matrix metalloproteinases, and other factors, which can support neoplastic progression. To investigate whether the inflammatory response to tumor implantation was affected by high-dose E2 treatment, we measured the number of TAMs by staining paraffin-embedded tumor sections with F4/80 antibody, which stains murine macrophages.

Based on the F4/80 staining observed in these sections, we observed a decreased number of macrophages at the periphery of E2-treated tumors. To quantify peripheral macrophages, we created a composite image of the entire tumor. We show that total F4/80 staining in the periphery of control tumors is significantly greater than that of E2 tumor sections (P = 0.0281). Figure 5A (control) and Fig. 5B (E2 treated) are representative photomicrographs of F4/80 staining patterns, and Fig. 5C shows the quantification of median peripheral F4/80 staining for control and E2 groups. These data estimate the number of macrophages recruited to the periphery of the tumor site that have not necessarily been incorporated and/or coopted by the tumor. Our results suggest that E2 treatment alters macrophage recruitment to the tumor site in this model but not infiltration into the tumor.


Figure 5
View larger version (21K):
[in this window]
[in a new window]
  		Fig. 5. Peripheral macrophage staining correlates with increased tumor size in control versus E2-treated mice. Bright-field photomicrographs (x2 magnification) of an entire cross section of tumors from control (A) and E2-treated (B) mice stained with F4/80 antibody. Bar, 2,500 µm. Extended views were obtained by "stitching" several overlapping fields from each tumor. Sections were counterstained with hematoxylin. Pixels positive for the HRP staining product were detected using the MetaMorph color thresholding function and normalized to the total live tissue area of the tumor determined using the MetaMorph color thresholding function for hematoxylin staining product. The normalized data correspond to the number of peripheral macrophages in each tumor section. C, median peripheral F4/80 staining of control and E2-treated tumors. Statistically significant differences in median values were determined by Mann-Whitney test. *, P = 0.0281.

 
To address the issue of whether E2 affected incorporation of macrophages recruited to the tumor site, we examined 20 independent fields (x20 magnification) from sections of control and E2-treated tumors and quantified the F4/80 staining using MetaMorph image analysis. There were no significant differences in intratumoral F4/80 staining in tumor sections from control versus E2-pelleted mice (data not shown). These data indicate that E2 treatment did not affect the infiltration of macrophages into the tumor mass and consequently do not support the idea that E2-induced tumor growth was due to an increase in the number of intratumoral macrophages in E2-treated mice.

E2 suppresses intratumoral inflammatory cytokine expression. TAMs produce several inflammatory cytokines and chemokines, such as tumor necrosis factor-{alpha}, which can either promote or inhibit tumor growth depending on the tumor microenvironment. We screened for expression of a panel of cytokines and chemokines to determine if differences exist between tumors from control and E2-treated mice. Table 1 depicts the mean cytokine measurements from homogenates of four separate tumors from each treatment group. Only cytokines or chemokines that were different between control and E2-treated tumors are shown. E2 treatment was associated with a suppressed level of several inflammatory cytokines, including IL-1ß, IL-1{alpha}, IL-6, GM-CSF, and G-CSF. These data provide additional supporting evidence that E2 treatment inhibits the immune response in a manner consistent with the observed increased tumor size in E2-treated mice.


View this table:
[in this window]
[in a new window]
  		Table 1. Cytokine and chemokine levels in tumors from control versus E2-treated mice

 

   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
The xenograft model of ER-negative human breast cancer described in these studies provides a starting point for investigation of the potential modulation of the host immune system by E2 independent of direct effects on tumor growth. In this model, MDA-MB-468 cells implanted in C57BL/6/SCID mice form significantly larger tumors in E2-treated versus control group. The E2-induced tumor growth requires ER-{alpha} because estrogen-stimulated tumor growth was abrogated in ER-{alpha}KO mice, indicating that ER-{alpha} mediated the effect of estrogen on tumor growth. The C57BL/6 mouse strain is relatively resistant to tumor growth when compared with other mouse strains, including the BALB/c mouse strain, which is widely used in human breast cancer xenograft studies (24). The relatively high degree of tumor resistance observed for C57BL/6 mice is based, at least in part, on increased NK activity due to the expression of the NK-activating receptor Ly-49D (33, 34). In our model, depletion of NK cells in this strain of mice also results in a significantly larger tumor mass when compared with control animals, indicating that an active immune surveillance system is operational in these mice. Additionally, we and others have shown that in vivo E2 treatment of C57BL/6 mice suppresses NK activity (10, 14, 35). Suppression of NK activity may be indirectly involved in E2-induced increased tumor size; however, it is not likely that E2 inhibition of NK cell activity was directly responsible for the increased tumor size in E2-treated animals because ER-{alpha}KO mice do not show E2-induced tumor growth, although E2 inhibits NK activity in ER-{alpha}KO mice. These data support the work of Friedl and Jordan (21) with a different ER-negative breast cancer cell line, MDA-MB-231 subclone 10A. This cell line xenografted into nude/beige mice that lack NK cells exhibited E2-stimulated growth, and therefore, they concluded that the observed E2-stimulated growth was not immune mediated. However, these authors did not rule out the possibility that other innate immune cells mediated the effect of estrogen in this model.

The hypothesis that tumors are recognized by the innate and adaptive immune systems is supported by recent studies showing an increased tumor incidence in animal and human immunocompromised hosts and by studies describing mechanisms used by tumors to escape detection by the immune system (reviewed in refs. 36–38). To study host, tumor, or environmental factors that facilitate suppression or immune escape by the tumor, a model system whereby active immune surveillance of tumor initiation and growth occurs is needed. As SCID mice lack functional T and B cells, tumor growth in this model is predominately controlled by cellular components of the innate immune system. The attendant inflammatory response to dysregulated and/or cancerous tissue growth involves not only NK cells but also many other immune cell subtypes, including macrophages, neutrophils, dendritic cells, etc. (reviewed in refs. 32, 39).

The relationship of the inflammatory response and cancer was hypothesized as early as 1863 by Virchow, and this connection has fostered the analogy that cancer is a "wound that does not heal" (40, 41). Interestingly, supplemental estrogens have been purported to have anti-inflammatory properties and promote wound healing in elderly males and females at least in part by regulating macrophage migration inhibitory factor (42, 43). In the rat brain, estrogen treatment reduced superoxide release and phagocytosis by microglial cells (44). In contrast, estrogen was reported to have proinflammatory effects on neutrophil and macrophage infiltration in the mouse uterus during the estrous cycle (45) and to induce proinflammatory cytokine production and subsequent inflammation in the rat prostate (46). Thus, tissue-specific interactions may drive the outcome of the response. In our model, estrogen treatment promoted significant growth of MDA-MB-468 tumors in C57BL/6/SCID mice, which our data suggest was due in part to inhibition of macrophage recruitment to the tumor site and alteration of inflammatory cytokine production. Although inhibition of macrophage recruitment is a potential mechanism through which estrogen inhibits tumor growth, other factors, such as macrophage-mediated cytotoxicity or cytokine production, may also mediate estrogen-induced tumor growth in this model. Production of proinflammatory cytokines, including IL-1ß, IL-1{alpha}, and G-CSF, in tumor tissues was decreased by estrogen treatment. These studies did not elaborate whether TAMs or the tumor cells were affected by the E2 treatment; however, because MDA-MB-468 cells are ER negative, E2 would likely be acting in a paracrine manner to modulate cytokine production by tumor tissue.

Macrophages and accompanying cytokine/chemokine production are necessary for normal mammary gland development, and macrophages have been implicated in breast cancer progression (31, 47, 48). However, the role of TAMs in etiology of breast and other cancers has been controversial. For instance, Lin and Pollard (31) crossed CSF-1-deficient mice (CSF–/–) that lack macrophages with mice expressing polyoma middle T oncoprotein, which spontaneously develop syngeneic mammary tumors. CSF–/–/polyoma middle T oncoprotein mice formed tumors that grew at a similar rate to polyoma middle T oncoprotein mice but progressed to a metastatic state at a slower rate, suggesting that TAMs promote breast cancer progression. These authors proposed that experimental models in which TAMs function to reject tumors are based on xenograft studies and may "simply reflect a host-versus-graft response." However, they also propose that rejection of xenografted tumors is based on antigen presentation to T cells. In our studies, we cannot eliminate a host-versus-graft reactivity, but T-cell involvement is not a factor in this model on a SCID background.

In summary, our data support the hypothesis that estrogens can promote growth of an ER-negative tumor by influencing the host immune response or tumor surveillance. Exposure to estrogens is considered a risk factor for the development of breast cancer, and our data indicate that the well-documented effects of E2 on immune function may play a role in this process. Specifically, we propose that E2 negatively affects the inflammatory response to the tumor. This model system may prove useful in delineating the process whereby tumors evade host immunity and even coopt the immune responses to promote their own growth. Ultimately, targeting the inflammatory response has been proposed as an immunotherapy for cancer (49). E2 modulation of the inflammatory response to tumor growth suggests that designing selective ER modulators specifically affecting inflammation may have a therapeutic value on immune surveillance of breast cancers, even ER-negative breast cancers.


   Footnotes
 
Grant support: Department of Defense Breast Cancer Research Program, Project No. DAMD17-03-1-0248.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 5/20/05; revised 6/13/06; accepted 6/22/06.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Henderson BE, Ross R, Bernstein L. Estrogens as a cause of human cancer: the Richard and Hinda Rosenthal Foundation award lecture. Cancer Res 1988;48:246–53.[Free Full Text]
  2. Marshall E. Epidemiology. Search for a killer: focus shifts from fat to hormones. Science 1993;259:618–21.[CrossRef][Medline]
  3. Clarke R. Human breast cancer cell line xenografts as models of breast cancer. The immunobiologies of recipient mice and the characteristics of several tumorigenic cell lines. Breast Cancer Res Treat 1996;39:69–86.[CrossRef][Medline]
  4. Engel LW, Young NA. Human breast carcinoma cells in continuous culture: a review. Cancer Res 1978;38:4327–39.
  5. Dickson RB, Lippman ME. Growth factors in breast cancer. Endocr Rev 1995;16:559–89.[CrossRef][Medline]
  6. Yager JD, Liehr JG. Molecular mechanisms of estrogen carcinogenesis. Annu Rev Pharmacol Toxicol 1996;36:203–32.[CrossRef][Medline]
  7. Ciocca DR, Roig LM. Estrogen receptors in human nontarget tissues: biological and clinical implications. Endocr Rev 1995;16:35–62.[CrossRef][Medline]
  8. Miller L, Hunt JS. Sex steroid hormones and macrophage function. Life Sci 1996;59:1–14.[CrossRef][Medline]
  9. Olsen NJ, Kovacs WJ. Gonadal steroids and immunity. Endocr Rev 1996;17:369–84.[CrossRef][Medline]
  10. Curran EM, Berghaus LJ, Vernetti NJ, Saporita AJ, Lubahn DB, Estes DM. Natural killer cells express estrogen receptor-{alpha} and estrogen receptor-ß and can respond to estrogen via a non-estrogen receptor-{alpha}-mediated pathway. Cell Immunol 2001;214:12–20.[CrossRef][Medline]
  11. Molero L, Garcia-Duran M, Diaz-Recasens J, Rico L, Casado S, Lopez-Farre A. Expression of estrogen receptor subtypes and neuronal nitric oxide synthase in neutrophils from women and men: regulation by estrogen. Cardiovasc Res 2002;56:43–51.[Medline]
  12. Lambert KC, Curran EM, Judy BM, Lubahn DB, Estes DM. Estrogen receptor-{alpha} deficiency promotes increased TNF-{alpha} secretion and bacterial killing by murine macrophages in response to microbial stimuli in vitro. J Leukoc Biol 2004;75:1166–72.[Abstract/Free Full Text]
  13. Sapino A, Cassoni P, Ferrero E, et al. Estrogen receptor {alpha} is a novel marker expressed by follicular dendritic cells in lymph nodes and tumor-associated lymphoid infiltrates. Am J Pathol 2003;163:1313–20.[Abstract/Free Full Text]
  14. Baral E, Nagy E, Berczi I. Modulation of natural killer cell-mediated cytotoxicity by tamoxifen and estradiol. Cancer 1995;75:591–9.[CrossRef][Medline]
  15. Baral E, Nagy E, Krepart GV, Lotocki RJ, Unruh HW, Berczi I. Antiestrogens sensitize human ovarian and lung carcinomas for lysis by autologous killer cells. Anticancer Res 2000;20:2027–31.[Medline]
  16. Baral E, Nagy E, Kwok S, McNicol A, Gerrard J, Berczi I. Suppression of lymphocyte mitogenesis by tamoxifen: studies on protein kinase C, calmodulin, and calcium. Neuroimmunomodulation 2000;7:68–76.[CrossRef][Medline]
  17. Screpanti I, Felli MP, Toniato E, et al. Enhancement of natural-killer-cell susceptibility of human breast-cancer cells by estradiol and v-Ha-ras oncogene. Int J Cancer 1991;47:445–9.[Medline]
  18. Berry J, Green BJ, Matheson DS. Modulation of natural killer cell activity by tamoxifen in stage I post-menopausal breast cancer. Eur J Cancer Clin Oncol 1987;23:517–20.[CrossRef][Medline]
  19. Brenner BG, Friedman G, Margolese RG. The relationship of clinical status and therapeutic modality to natural killer cell activity in human breast cancer. Cancer 1985;56:1543–8.[CrossRef][Medline]
  20. Swain SM. Tamoxifen for patients with estrogen receptor-negative breast cancer. J Clin Oncol 2001;19:93–7S.
  21. Friedl A, Jordan VC. Oestradiol stimulates growth of oestrogen receptor-negative MDA-MB-231 breast cancer cells in immunodeficient mice by reducing cell loss. Eur J Cancer 1994;30A:1559–64.[CrossRef]
  22. Welshons WV, Grady LH, Engler KS, Judy BM. Control of proliferation of MCF-7 breast cancer cells in a commercial preparation of charcoal-stripped adult bovine serum. Breast Cancer Res Treat 1992;23:97–104.[CrossRef][Medline]
  23. Natarajan N, Shambaugh GE 3rd, Elseth KM, Haines GK, Radosevich JA. Adaptation of the diphenylamine (DPA) assay to a 96-well plate tissue culture format and comparison with the MTT assay. Biotechniques 1994;17:166–71.[Medline]
  24. Maruo K, Ueyama Y, Hioki K, Saito M, Nomura T, Tamaoki N. Strain-dependent growth of a human carcinoma in nude mice with different genetic backgrounds: selection of nude mouse strains useful for anticancer agent screening system. Exp Cell Biol 1982;50:115–9.[Medline]
  25. Hrushesky WJ, Gruber SA, Sothern RB, et al. Natural killer cell activity: age, estrous- and circadian-stage dependence, and inverse correlation with metastatic potential. J Natl Cancer Inst 1988;80:1232–7.[Abstract/Free Full Text]
  26. Tsai EM, Wang SC, Lee JN, Hung MC. Akt activation by estrogen in estrogen receptor-negative breast cancer cells. Cancer Res 2001;61:8390–2.[Abstract/Free Full Text]
  27. Rubanyi GM, Johns A, Kauser K. Effect of estrogen on endothelial function and angiogenesis. Vascul Pharmacol 2002;38:89–98.[CrossRef][Medline]
  28. Dabrosin C, Palmer K, Muller WJ, Gauldie J. Estradiol promotes growth and angiogenesis in polyoma middle T transgenic mouse mammary tumor explants. Breast Cancer Res Treat 2003;78:1–6.[CrossRef][Medline]
  29. Dabrosin C, Margetts PJ, Gauldie J. Estradiol increases extracellular levels of vascular endothelial growth factor in vivo in murine mammary cancer. Int J Cancer 2003;107:535–40.[CrossRef][Medline]
  30. Leek RD, Harris AL. Tumor-associated macrophages in breast cancer. J Mammary Gland Biol Neoplasia 2002;7:177–89.[CrossRef][Medline]
  31. Lin EY, Pollard JW. Macrophages: modulators of breast cancer progression. Novartis Found Symp 2004;256:158–68.[Medline]
  32. Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420:860–7.[CrossRef][Medline]
  33. Nakamura MC, Naper C, Niemi EC, et al. Natural killing of xenogeneic cells mediated by the mouse Ly-49D receptor. J Immunol 1999;163:4694–700.[Abstract/Free Full Text]
  34. Idris AH, Smith HR, Mason LH, Ortaldo JR, Scalzo AA, Yokoyama WM. The natural killer gene complex genetic locus Chok encodes Ly-49D, a target recognition receptor that activates natural killing. Proc Nat Acad Sci U S A 1999;96:6330–5.[Abstract/Free Full Text]
  35. Seaman WE, Merigan TC, Talal N. Natural killing in estrogen-treated mice responds poorly to poly I.C despite normal stimulation of circulating interferon. J Immunol 1979;123:2903–5.[Abstract/Free Full Text]
  36. Pardoll D. Does the immune system see tumors as foreign or self? Annu Rev Immunol 2003;21:807–39.[CrossRef][Medline]
  37. Smyth MJ, Hayakawa Y, Takeda K, Yagita H. New aspects of natural-killer-cell surveillance and therapy of cancer. Nat Rev Cancer 2002;2:850–61.[CrossRef][Medline]
  38. Vicari AP, Caux C, Trinchieri G. Tumour escape from immune surveillance through dendritic cell inactivation. Semin Cancer Biol 2002;12:33–42.[CrossRef][Medline]
  39. Zitvogel L. Dendritic and natural killer cells cooperate in the control/switch of innate immunity. J Exp Med 2002;195:F9–14.
  40. Balkwill F, Mantovani A. Inflammation and cancer: back to Virchow? Lancet 2001;357:539–45.[CrossRef][Medline]
  41. Dvorak HF. Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. N Engl J Med 1986;315:1650–9.[Medline]
  42. Ashcroft GS, Ashworth JJ. Potential role of estrogens in wound healing. Am J Clin Dermatol 2003;4:737–43.[CrossRef][Medline]
  43. Ashcroft GS, Mills SJ, Lei K, et al. Estrogen modulates cutaneous wound healing by downregulating macrophage migration inhibitory factor. J Clin Invest 2003;111:1309–18.[CrossRef][Medline]
  44. Bruce-Keller AJ, Keeling JL, Keller JN, Huang FF, Camondola S, Mattson MP. Antiinflammatory effects of estrogen on microglial activation. Endocrinology 2000;141:3646–56.[Abstract/Free Full Text]
  45. Tibbetts TA, Conneely OM, O'Malley BW. Progesterone via its receptor antagonizes the pro-inflammatory activity of estrogen in the mouse uterus. Biol Reprod 1999;60:1158–65.[Abstract/Free Full Text]
  46. Harris MT, Feldberg RS, Lau KM, Lazarus NH, Cochrane DE. Expression of proinflammatory genes during estrogen-induced inflammation of the rat prostate. Prostate 2000;44:19–25.[CrossRef][Medline]
  47. Gouon-Evans V, Lin EY, Pollard JW. Requirement of macrophages and eosinophils and their cytokines/chemokines for mammary gland development. Breast Cancer Res 2002;4:155–64.[CrossRef][Medline]
  48. Gouon-Evans V, Rothenberg ME, Pollard JW. Postnatal mammary gland development requires macrophages and eosinophils. Development 2000;127:2269–82.[Abstract]
  49. Wahl LM, Kleinman HK. Tumor-associated macrophages as targets for cancer therapy. J Natl Cancer Inst 1998;90:1583–4.[Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
1078-0432.CCR-05-1117v1
12/19/5641    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Curran, E. M.
Right arrow Articles by Estes, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Curran, E. M.
Right arrow Articles by Estes, D. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online